Thursday, January 22, 2009

Logan's Run: The Finish Line

My differences with the classification and ranking system isn't that the concept is defective but that the present implementation is faulty. Clearly the PSP disagrees and beyond that has a purpose in the way the system currently functions (and will purportedly function in the future.) My objection is two-fold; the current system is unnecessarily coercive and, in the end it won't work as intended. First thing I looked at was the points system (which isn't bad but in some respects is a blunt instrument.) It would be possible to add to the present calculations and it would be possible to put a different points system in place but how the numbers are figured isn't the real problem. And making that change wouldn't automatically fix things. So while I think there's some room for improvement there it's not a big deal.
As discussed in 'The Pear or the Pyramid' post one of my objections is demographic. I don't think you can re-make the ranks to reflect what you want them to look like without losing teams & players if what you want is 'unnatural'. It appears to me the idea is to remedy the dearth of upper division teams by redistributing the lower division teams in the sense that what we now consider D2 level talent becomes D1 and so on. Which also begs the question what happens to legitimate semi-pro talent when their division is populated by D1 talent? Or is that just a big mixer? It also presupposes the plan will work but that assumes there otherwise isn't enough real D1/semi-pro talent out there and it also assumes that there are no other causal factors inhibiting teams at the upper levels. My view is there are substantial causal factors inhibiting upper level team growth and dumbing down the divisions solves nothing. (And, of course there are plenty of D1/Pro (semi-pro) ranked players in the system who aren't competing so the issue isn't lack of players.)
I also think some of the changes made to the system this year tacitly acknowledge my views. The (marginal) relaxation of D3 classification rules and the inclusion of D3/D4 Intro are plainly moves made to encourage more D3 level participation both by "new" teams and by D3 teams that may have intentionally limited their participation in the past to avoid being moved up to D2. And as soon as you accept this premise it must follow that it applies, to one degree or another, across all the divisions. [This also begins to support the idea that coercive measures are in some respects counterproductive.]
Even if one were to argue that self-selection (teams choosing to participate) tends to skew the expected numbers towards a higher baseline of ability across the board all the limited numbers available suggest that doesn't really matter. A quick examination of where teams place themselves produces the (broadly) expected ratios, ie; they's mostly D2 & D3.
My objection to coercion is practical. It's simply not terribly effective and when coupled with a restructuring policy that will not work cannot achieve its intended result.
Another objection I have to strip mining the (presumed) upper ranks of the lower divisions is that it offers no continuity, no normative standard to be achieved or measured against (and understood) by the incoming teams. For example, if the top 4 teams are moved up the expectation is that the teams that finished 5 - 8 should be at or near the top the following season. This won't be universally true but it gives all the other teams a way of measuring their progress and/or ability.

There are two options in dealing with the present system; simply reign in its excesses and move up only those players/teams that must move in order to sustain the competitive balance of the division or, include a more flexible means for players to move down as well as up in order to be responsive to some of the current system's excesses. The first option is easiest but undercuts the PSP's restructuring of divisions plan. The second is harder to order by rule but can be done. (I don't know what impact it would have given the current system but in any system it should provide greater opportunity without disrupting competitive balance any more than occurs now with the allowances of limited upper division players on lower division rosters.)
For example, a simple change might be – every team that wins an event moves up and in addition any team that has more than one top 4 finish moves up. For '09 that would still move up 6 teams in D3 and 7 teams in D2 but it would only be winning teams and/or teams that showed the consistent ability to finish near the top. (Btw, it will be interesting to see how many of the D2 teams forced to move up actually ever play a D1 event. The going rate recently is about 50% which also seems to fly in the face of the PSP's goals here.)
As for players dropping ranks I have worked out a system. It has a pro floor and a floor for all other players. It takes into account the "success" of players in the ranks and has measures to limit possible abuses. It is predicated on my belief that some measure of flexibility with respect to rank will allow wider participation without harming competitive balance by keeping more players actively in the mix of pursuing competitive opportunities at appropriate levels of play. (Should any established league or ranking system have an interest I'll be happy to provide the details.)

The alternative is a completely redesigned classification & ranking system. Even so, it would necessarily look something like the current one and still wouldn't be able to address the populations of the divisions that the PSP is apparently trying to manipulate with the present system. In order to "fix" divisional disparities one must do two basic things; identify the causes and determine a methodology for dealing with the problem. For example, the PSP sees a dearth of upper division teams and their methodology is coercion and their identified cause is not enough upper division teams (which is a tautology) and their fix is to turn D2 teams into D1 teams and presto, plenty of d1 teams. (Or maybe not.)
My view is classification and ranking can't address the issue alone. Classification and ranking is only about the competitive integrity of the divisions. Other concerns require other answers. If coercion isn't the answer, what is? I'm thinking incentives. Give teams a reason or reasons to do what you'd like them to do. Here's a start: tie entry fees to paintball played. Since D2 and D1 now both play Race 2-5 their entries should be the same. Now there are no series prizes. Change that and award a D1 series prize package along with event prizes. These changes begin to reward excellence (higher division play) and motivate teams to move up if they think they can compete and at the same time removes the disincentive of higher entry fees for higher division play. There are, I'm sure, plenty of creative ways to achieve the desired goals. My choice would be to institute elements of all the options mentioned.

There you have it, kids.

14 comments:

Joe said...

So, if there were an actual way to track player performance on field (strictly speaking non-5 man format here), such as the basic stat tracking done by the NXL, do you think that could in some way be used to also determine the classification of a player?

For example, a player that has played 5 events as D3, but is always on the field when their team loses a point, serves penalty times, and frequently gets eliminated (>75% of the points they've played), would stay ranked D3 despite the fact that other players, or even his team , has been bumped up to D2.

This may be impractical currently as the league would have to employ statisticians at each field, but should a way be determined to track such info, would it alleviate some of your concerns?

D.MaShadow said...

Ha So true! I'd suggest that whole win an event you get to bump divisions but the only problem with that might be some teams will play the first event and the one team that happens to win. then doesn't play the rest of the events. That's probably the only problem see with that .. even though I think that whole idea you mentioned about the bumping up and stuff is a great idea.

Baca Loco said...

Hey Joe
It's an interesting idea but you're also correct in thinking it's not a realistic option. Even in last year's NXL they didn't always do a very good job and even with the data they accumulated they didn't examine it in the detail you suggest might be helpful.
In the players' interest I'd be satisfied with using the classification system to only move up the best teams and if you added some flexibility to the classification system so players could move down as well as up I think it would address most of the problems.
In the league's interest I would like to see positive moves made to help create and support healthy divisions up and down the ladder. And something I didn't address is how, outside of classification and ranking, the league looks at the thin upper divisions because I think it's going to take a lot more than trying to turn D1 into the new D2.

D.Mash
Fierce Army did almost exactly that only playing (and winning) the final event. So, yes that sort of thing could happen and it's conceivable it was a huge fluke--but probably not. And it's a lot better to my way of thinking than moving up teams that never finished in the top 4 but played all or most of the events. I don't care how many you play if you demonstrate D2 level ability then you ought to stay D2 and there is nothing wrong with that. My only reservation is that different events have different pools of teams competing and are consequently not identical and frequently not close to identical. But I don't see a way around that and I tried a number of ways to adjust for differing pool size.

Lawrence said...

http://www.paintball-players.org/UniversalClassificationProgramDRAFT.pdf

how interesting...check the GOALS section.

Baca Loco said...

Lawrence
If you didn't see them at the time you might want to take a look at the 'Brave New Paintball World' posts end of August, first part of September.
Saw this recently and privately offered an opinion. Now that I see the context I'll have more over the weekend--though maybe not till Sunday what with practice and the Affliction show Satruday night. :-)

Lawrence said...

yes, that's why i posted it here -- the parallels are striking!

:o
LoL!

raehl said...

I don't think the goal of the classification system is to populate the upper divisions. The goal of the classification system is twofold:

First, provide a place for players who don't know WTF they are doing on a logistical level a place to play. (I.e., before you go play big-stakes poker, you need to learn the rules and the basic flow of the game.) That's what the entry level divisions are for, and everybody should be kicked out of them fairly quickly, because no matter how bad you might suck, you still know the rules, and the entry-level division is for people who don't know the basics of the format.

Then, once you're past knowing the basics, the goal becomes making teams in each division competitive with one another. Competitive doesn't mean the best team is good. Competitive means that the best team and the worst team can have a match that is a contest. Clearly, last season, D1 absolutely failed at being competitive. And to an extent, D2 did as well.

But no matter how you slice it, some players are better than others. So what did we do with PSP teams this year?

The group of teams who were winning D1 by a large margin became Semi-Pro. The group of teams who were winning D2 got combined with the bottom of D1 to form the new D1. The remainder of D2 and the top of D3 get to be the new D2, and the teams that just plain do not perform are the new D3, with a NEW Intro division to cover the need for a place where teams can get acclimated to the logistics of playing.


So, here is the question: Are the players in each division of a similar skill level? And the answer is, YES, ABSOLUTELY. I can not think of any better way to get people of similar skill levels in the same division.

Are the best 10% of national-level XBall players classified Pro? Yes. Next 15% in Semi-Pro? Yes. Next 20% in D1? Yes. Next 25% in D2? Yes. Next 30% in D3? Yes. Are those divisions thus similarly skilled players who will form competitive teams? Yes.

And that's all we can ask of the classification system - put players of a similar skill level in the same class. Teams not attending, or players not having a place to play, are NOT CLASSIFICATION PROBLEMS! And the solution to something that is not a classification problem is NOT to say "Oh, you can't find a D1 team to play on? Well, shoot, let's just put 45% of players in D2."

Does moving people up cause some players to stop playing? Maybe. But not moving them up is worse.

By the way, 50% of players staying to play D1 is *GOOD*. Well, the leaving isn't good, but the attrition rate isn't bad, as 50% of national-level paintball players leave every season anyway, whether you move them up or not. You're just not paying attention to the ones who don't get moved up and still leave.

raehl said...

Short comment:

If a team stops playing because they get moved up to D1 from D2, there is only one reason: They don't want to play if they only get to play people as good as they are. So if they get moved up, and stop playing, either they were going to stop playing anyway (finances etc) or they just didn't want to play if they actually have to play in a competitive division. And we can't base classification rules around teams not wanting to actually compete.

Baca Loco said...

Chris,
Slow start but you picked up a little steam eventually. Unfortunately I can't respond right now but I'll come back to it.
Here's a hint though--you undercut part of your own argument in the beginning and you and I both know full well part of the current classification process is intended to redistribute players in such a way as to alter the nature of the divisions. Now if you want to argue that despite that the result is or will be as you say that's fine.

Joe said...

I think, and I may be a bit off kilter here, but we need to define a difference between classification and ranking. It seems to me that the two terms are used synonymously when they actually mean two different things.

A player's classification should be determined based on their national/regional experience. Essentially, their classification is how their class/ranking is determined now, based on events and points accumulated. An individual players ranking should be based relative to their peers, which is something that we have been unsuccessful in tracking, as we have currently no means to do so. Mostly, this is where individual on-field participation comes into play.

raehl said...

Baca,

What do you mean by "alter the nature of the divisions"? If you mean alter them so that players in each division are of a similar skill level, then yes, that's exactly what the classification system is designed to do. 40% of the players in one division and 40% of the players in another division does NOT yield competitive divisions, and the classification system absolutely tries to better group players by skill/talent.

Baca Loco said...

Joe,
Realistically there's only so much any system can do. The best one can hope for is competitive balance within the divisions while classifying players as fairly as possible.
Ranking is simply the method used as it serves to rank teams relative to their competition.
And of course that is not really what the PSP system is doing. It is trying to restructure the divisions in such a way that the preponderance of the ability in the future D1 is the equivalent of what is currently in D2.

Chris
And that is what I meant by altering the nature of the divisions. The current system is intentionally altering the range of ability that qualifies for that division. There is a substantive difference between a competitive division in isolation and a system that on the whole provides competitive balance AND is as widely inclusive up and down the divisions as possible in order to encourage the widest possible participation.
I don't care if you guys want to stuff semi-pro with D3 quality players--if they're all the same it's a competitive division. But doing stuff like that causes other problems.
You keep wanting to fight the battle over is the current system ok or not. That's not really my issue with it. My issue with it is that it needs to be better or different BECAUSE this, this and this are happening as a result. Not because it could be better.

I'll be posting on the new universal classification gimmick tomorrow (hopefully) and if I've failed to address any of your positions vis-a-vis classification/ranking let me know.

raehl said...

So maybe I missed it, but did you offer any BETTER alternative than the current system?

I don't think anybody is saying what we have is perfect, or necessarily even close to it, but nobody seems to have anything better either.

Baca Loco said...

Reading comprehension, Chris, c'mon. I'm not dissing APPA or even the ranking system (which is ok if not ideal.) I am principally opposed to the classification rules and I suggested a variety of alternatives.
And you couldn't wait until I posted tomorrow (later today, could you?)